On 6/5/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Luca Tettamanti wrote:
> > Hello,
> > I'm testing the new Fedora7 under KVM. As you may know Fedora has
> > migrated to the new libata drivers.
> >
> > ata_piix is unhappy with
On 6/5/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Luca Tettamanti wrote:
> > Il Mon, Jun 04, 2007 at 11:51:10PM +0300, Avi Kivity ha scritto:
> >
> >>> While doing repeated tests with the installer I ran into another
> >>> (unrelated) problem
On 6/10/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Luca wrote:
> > On 6/5/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> >> Luca Tettamanti wrote:
> >> > Il Mon, Jun 04, 2007 at 11:51:10PM +0300, Avi Kivity ha scritto:
> >> >
> >> >&
On 6/11/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Luca wrote:
> >>
> >> I've managed to reproduce this on kvm-21 (it takes many boots for this
> >> to happen, but it does eventually).
> >
> > Hum, any clue on the cause?
>
> From what
ght on the lockup. If nothing happens
it means that CPU is not even receiving NMIs...
Luca
-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of yo
On 6/15/07, Luca Tettamanti <[EMAIL PROTECTED]> wrote:
> Il Fri, Jun 15, 2007 at 12:53:24AM +0200, Luca Tettamanti ha scritto:
> > Il Thu, Jun 14, 2007 at 11:26:29AM +0300, Avi Kivity ha scritto:
> > > Luca Tettamanti wrote:
> > > >With GOOD_APIC apic_read_arou
use to unload
> > (or there's a nasty bug somewhere).
>
> Nahh, that's too brutal. Is there a way to (hypothetically) read:
>
> # cat /sys/module/kvm/parameters/number_of_running_vms
> 2
cat /sys/module/kvm
ng to GIT those files haven't been touched, something went
wrong during the initial import.
Luca
-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of
pe
> on the keyboard while the kernel boots (before the actual hang). Seems to be
> very random though.
Try booting with 'nolapic' on kernel command line.
Luca
-
This SF.net email is sponsored by DB2 Express
Do
tion (the device *always* uses a write
transaction to signal the interrupt condition, and the host bridge may
turn the message into a legacy interrupt) so that the OS can drive the
device even without fiddling with MSI.
As Eddie said XP probably would be confused by MSI-only devices.
Luca
On 6/22/07, Luca <[EMAIL PROTECTED]> wrote:
> On 6/22/07, Morten Lund <[EMAIL PROTECTED]> wrote:
> > On Thursday 21 June 2007 01:24, Luca wrote:
> > > On 6/20/07, SourceForge.net <[EMAIL PROTECTED]> wrote:
> > > > Bugs item #1740031, was opened a
On 7/2/07, Mark Ryden <[EMAIL PROTECTED]> wrote:
> My question is:
> I am launching windows 2003 from linux by
> "/usr/local/kvm/bin/qemu-system-x86_64 . -hda vdisk.img -cdrom
> /dev/hdb -no-acpi -boot d -m 384".
-full-screen and, please, man pages
> p = mc->objects[--mc->nobjs];
> memset(p, 0, size);
> return p;
> }
MMU working memory was exhausted during a guest context switch. It has
been fixed by:
KVM: Lazy guest cr3 switching
rse, but that doesn't save you from running a local SPAM filter.
FWIW GMail spam filter catches almost all the spam on this list,
spamassassin on my machine does the rest.
Luca
-
This SF.net email is sponsored by DB2 Ex
on X86 && EXPERIMENTAL
> + depends on X86 && EXPERIMENTAL && PREEMPT_HOOKS
> depends on X86_CMPXCHG64 || 64BIT
select PREEMPT_HOOKS maybe?
Luca
-
This SF.net email is sponsored by DB2 Exp
unique net segment
I have a script for the 2nd solution, I can send as soon as I get home
if you're interested.
Luca
-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express
mmit id by chance?
> >
> > -- Dave
> >
> >
> commit 55a3212bc2f5ecddcd4c5cdf2bfb37ad71e45ff2
> Author: Luca Tettamanti <[EMAIL PROTECTED]>
> Date: Tue Jun 5 14:47:33 2007 +0300
>
> kvm: qemu: initialize ata ports as enabled
>
> this allows libata to see the qemu
uest
> would fail to see the IDE controller. I found it to be that the IDE
> controller wasn't enabled sometimes and setting the 0x80 bits as Luca
> has done fixed the issue.
>
> Then, I found that later versions of KVM (sans my patch) seemed to work
> fine so I somewhat assume
On 7/13/07, Gregory Haskins <[EMAIL PROTECTED]> wrote:
> On Fri, 2007-07-13 at 16:46 +0200, Luca wrote:
> > On 7/13/07, Gregory Haskins <[EMAIL PROTECTED]> wrote:
>
> > Just to clarify: you are suggesting that the "old" IDE driver used to
> > see tha
;
> There's your problem. With that patch, your kernel is using PIO instead
> of DMA.
>
> I'm not 100% sure what's going here other than the fact that this
> particular problem has cropped up a number of times on qemu-devel. It's
> been very difficult to g
t 0x1408-0x140f, BIOS settings: hdc:pio, hdd:pio
Hum, FC6 PIIX driver always comes up with the same mode, regardless of
what I'm doing to the controller...
Luca
-
This SF.net email is sponsored by DB2 Express
Download DB2 Exp
On 7/15/07, Luca <[EMAIL PROTECTED]> wrote:
> On 7/13/07, Dave Hansen <[EMAIL PROTECTED]> wrote:
> > diff -ru kvm-fast-dmesg.txt kvm-slow-dmesg.txt
> >
> > Linux version 2.6.22 ([EMAIL PROTECTED]) (gcc version 4.1.2 (Ubuntu
> > 4.1.2-0ubuntu4)) #13 Wed
On 7/15/07, Luca <[EMAIL PROTECTED]> wrote:
> On 7/15/07, Luca <[EMAIL PROTECTED]> wrote:
> > On 7/13/07, Dave Hansen <[EMAIL PROTECTED]> wrote:
> > > diff -ru kvm-fast-dmesg.txt kvm-slow-dmesg.txt
> > >
> > > Linux version 2.6.22 ([EMA
On 7/16/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Luca wrote:
> > On 7/15/07, Luca <[EMAIL PROTECTED]> wrote:
> >> On 7/15/07, Luca <[EMAIL PROTECTED]> wrote:
> >> > On 7/13/07, Dave Hansen <[EMAIL PROTECTED]> wrote:
&
le (the
same way you do with real hw). Append "console=ttyS0" (or
"console=tty0 console=ttyS0" to have both) to kernel command line.
Luca
-
This SF.net email is sponsored by DB2 Express
Download DB2 Express C
On 7/17/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Luca wrote:
> >>
> >> So, would it be correct to revert the patch enabling the ports, assuming
> >> people use the kvm-supplied bios?
> >
> > I think so.
>
> Please confirm that the original
S (e.g. QEMU BIOS) which
doesn't enable the port though.
Luca
As a side note: I'm currently busy with ICWE and tomorrow I'll be
leaving for Rome so I cannot do further tests.
-
This SF.net email is spons
HD in win xp it also indictates 6.4GB.
Hum, then the used disk space is 6.4GB. "Selecting all files" may skip
hidden/system/whatever files according to explorer settings.
Luca
-
This SF.net email is sponsored by: Splunk
ly on of them can deliver the periodical interrupt. I
think that work to program the HPET in dont-break-other-stuff mode is
underway.
Luca
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files
On 8/7/07, Dor Laor <[EMAIL PROTECTED]> wrote:
> Luca claims the HPET intefer the RTC. Can it be disabled? ( I know some
> new chipsets implement rtc using HPET).
Basically HPET can operate in legacy mode - where it uses the same IRQ
as the RTC (and RTC won't deliver any i
pect that we just have a bad ISP that doesn't peer well. We'll
> work to fix this.
Seems more like a firewall or something like that. IP routing is ok, I
can reach the server (from Italy) just fine, but nothing responds
(nothing at *all*, not event a RST - the same behaviour of i
On 8/13/07, Daniel P. Berrange <[EMAIL PROTECTED]> wrote:
> On Mon, Aug 13, 2007 at 11:04:46AM +0300, Avi Kivity wrote:
> > Luca Tettamanti wrote:
> > Something like:
> >
> > - try to use HPET (unless -no-rtc selected)
> > - try to use RTC (unless -no-rtc
support
> I ran /usr/bin/qemu as I didn't have /usr/local/kvm/bin/qemu like the HOWTO
> mentioned (as I didn't compiled the kvm bundle myself...).
Under debian the executable is called kvm; I believe that it's the
same under
ed on to buidling
> kvm-33 from source tar ball which built fine. But, the kvm-33 tarball
> had the svm source files in the the kernel directory. I notice those
> files are not in the kvm-35 kernel directory (and are causing the
> failure above). Is there some other repo svm files a
On 8/13/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Luca Tettamanti wrote:
> > The patch stops the "unhandled wrmsr", but reboot is still not working
> > (guest is stuck using 100% of the CPU). The last working userspace is
> > KVM-28, and I tested it with re
the improvement in a linux guest
> when running on a windows host ?
The improvements - beyond the refactoring - are either specific to
Linux (HPET timer) or to UNIX in general (dynticks - POSIX timers are
used).
It may be possible to use one-shot timer on windows too, but I'm
iced I
> also needed):
>
> apt-get install kernel-package
> apt-get install linux-headers-2.6-amd64
Since Debian has a KVM package you can do:
apt-get build-dep kvm
which should cover all the dependencies.
Luca
rns I don't
> want to do.
>
> The error message is : "warning: could not configure /dev/net/tun: no
> virtual network emulation
> Could not initialize device 'tap''
You need CAP_NET_ADMIN to fiddle with TUN/TAP.
Luca
On 8/19/07, Eugene Coetzee <[EMAIL PROTECTED]> wrote:
> Luca wrote:
> >On 8/19/07, Eugene Coetzee <[EMAIL PROTECTED]> wrote:
> >>Hi .
> >>
> >>I'm trying to run KVM on Feisty, 64 bit Intel with the "-net nic -net
> >>tap" swi
On 8/19/07, Luca Tettamanti <[EMAIL PROTECTED]> wrote:
> +static uint64_t qemu_next_deadline(void) {
> +uint64_t nearest_delta_us = ULLONG_MAX;
> +uint64_t vmdelta_us;
Hum, I introduced a bug here... those vars should be signed.
On the overhead introduced: how do you me
On 8/20/07, malc <[EMAIL PROTECTED]> wrote:
> On Mon, 20 Aug 2007, Luca Tettamanti wrote:
>
> > Il Sun, Aug 19, 2007 at 10:31:26PM +0300, Avi Kivity ha scritto:
> >> Luca wrote:
> >>> On 8/19/07, Luca Tettamanti <[EMAIL PROTECTED]> wrote:
> >
On 8/21/07, Matthew Kent <[EMAIL PROTECTED]> wrote:
> On Sat, 2007-18-08 at 01:11 +0200, Luca Tettamanti wrote:
> > plain text document attachment (clock-hpet)
> > Linux operates the HPET timer in legacy replacement mode, which means that
> > the periodic interrupt of th
On 8/22/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Luca Tettamanti wrote:
> > Il Wed, Aug 22, 2007 at 08:02:07AM +0300, Avi Kivity ha scritto:
> >
> >> Luca Tettamanti wrote:
> >>
> >>
> >>> Actually I'm having troubles with cycleso
On 8/22/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Luca wrote:
> >>> This is QEMU, with dynticks and HPET:
> >>>
> >>> % time seconds usecs/call callserrors syscall
> >>> -- --- --- - --
On 8/22/07, Luca <[EMAIL PROTECTED]> wrote:
> I see a lot of sub ms timer_settime(). Many of them are the result of
> ->expire_time being less than the current qemu_get_clock().
False alarm, this was a bug in the debu
On 8/22/07, Luca <[EMAIL PROTECTED]> wrote:
> On 8/22/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> > Luca wrote:
> > >>> This is QEMU, with dynticks and HPET:
> > >>>
> > >&
thus can be
> >> ignored).
> >> VIRTUAL just becomes more accurate with dyntics, before multiple
> >> timers were batched together.
> >>
> >> > Or maybe your host kernel can't support such a high rate.
> >>
> >> I don'
be a 64bit integer (and the shift would be
acceptable).
There are a couple more of warning related to this issue, a possible
solution to silence the warning is to use always 64bit integers
instead of the longs.
Luca
-
This SF.n
On 8/23/07, Cam Macdonell <[EMAIL PROTECTED]> wrote:
> I'm trying to setup networking on FC6. Where can I find the tun/tap
> interface?
CONFIG_TUN
But I'm pretty sure that default FC kernel has the module already built. T
h -vnc :0 (or :1, :2, etc. - in this way you can manage
multiple independent VMs), then connect with vncviewer :0 (or
123.123.123.123:0 for remote connections).
Luca
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping
anilla 2.6.22.3 on that machine.
Try:
cat /sys/devices/system/clocksource/clocksource0/available_clocksource
do you see HPET listed twice?
Luca
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find
you run "make kernel V=1" (on a clean source) and post the result?
> Am I losing anything important by commenting them out?
You can workaround by adding:
#define KVM_EXIT_SET_TPR 11
Luca
-
This SF.net email is sp
usr/src/kernels/2.6.22.4- 65.fc7-x86_64/include -MMD -MF ./.kvmctl.d
> -g -fomit-frame-pointer -Wall -m64 -fno-stack-protector -c -o kvmctl.o
The include directory is wrong. Have you used --with-patched-kernel?
Luca
-
Thi
your own kernel modules. Do you
*really* want to do that?
Luca
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX
ied.
> if it isn't guaranteed to be the current thread, we're back
> with scheduler involvement, and possibly cacheline bouncing.
It's possible to use pthread_sigmask() to block the signal on all
threads but one. But this w
copies or substantial portions of the Software.
[cut]
Shouldn't you use the standard GPL boilerplate? (IANAL, so maybe the
text above is a superset of the GPL and is fine...)
Luca
-
This SF.net email is sponsored by: Splun
ong branch ;-) )
I prefer to clone the whole tree (disk space is cheap...), sharing GIT
object where possible, e.g.
git clone --reference kvm.avi
git://kvm.qumranet.com/home/dor/src/virtio/kvm kvm.dor
where kvm.avi is you existing kvm git repository.
Luca
On 8/29/07, Cam Macdonell <[EMAIL PROTECTED]> wrote:
> Luca wrote:
> > On 8/29/07, Cam Macdonell <[EMAIL PROTECTED]> wrote:
> >> Dor Laor wrote:
> >>> The code can now be accessed by git:
> >>> git://kvm.qumranet.com/home/dor/src/virtio/kvm
what can be a problem? is it a known bug or it's just not working for me?
It's known, but ATM there's no fix available. For more details see:
http://thread.gmane.org/gmane.comp.emulators.kvm.devel/5651/focus=6003
and following mails.
Luca
--
On 9/3/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Farkas Levente wrote:
> > Luca wrote:
> >
> >> On 9/3/07, Farkas Levente <[EMAIL PROTECTED]> wrote:
> >>
> >>> hi,
> >>> i've got a centos-5 x86_64 kvm-35 host system wit
Linux
may refuse to use ACPI without them (it depends on the configuration).
Luca
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio
aka "Pacifica".
> The $1M (or $0.02) question is... does KVM include any optimizations for the
> AMD virtualization extensions?
Yes, load kvm-amd.
Luca
-
This SF.net email is sponsored by: Microsoft
Defy a
On 9/8/07, Farkas Levente <[EMAIL PROTECTED]> wrote:
> Luca wrote:
> > On 9/8/07, Farkas Levente <[EMAIL PROTECTED]> wrote:
> >> kvm is not ready for production use for many reason:
> >> it can't reboot which imho a very basic feature, what's more c
Yet I visit some articles like this
> one
>
> http://aplawrence.com/Linux/kvm_virtualization.html
>
> which claims that Xen is "THE FASTEST" approach to virtualization. How can
> it be faster since it uses paravirtualization (software) instead of direct
> hardwa
On 9/9/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Luca Tettamanti wrote:
> > Hi,
> > while debugging a program with GDB I got a GPF which seems related to
> > KVM. Note that I was *not* using the debugger inside the guest nor on
> > the virtual machine. The
On 9/9/07, Anthony Liguori <[EMAIL PROTECTED]> wrote:
> Luca wrote:
> > On 9/9/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> >
> >> Luca Tettamanti wrote:
> >>
> >>> Hi,
> >>> while debugging a program with GDB I got a GPF which
> $dmesg
> [54716.482444] kvm: Unknown symbol __divdi3
kvm.ko fails to load due to a 64 bit division, see:
http://git.kernel.org/?p=linux/kernel/git/avi/kvm.git;a=commitdiff;h=e44af0f4ee99974ce40102e23784bc3cae7f4466
Luca
>
> [54716.492596] kvm_intel: Unknown symbol kvm_lapic_rese
gt; code: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 --> 00
> > 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> > 00 00 00 00 00
> > Cancelado (core dumped)
>
> Can you give us some additio
windows xp and got the following
> > output.
> >
>
> Reboot was improved in kvm-39, but not completely fixed. I'll nail it
> down in kvm-40 or kvm-41.
You can reboot linux guests passing "reboot=b" on kernel command line.
Don't know about win...
Luca
-
an you setup a serial console
(Documentation/serial-console.txt) or at least netconsole
(Documentation/networking/netconsole.txt)?
> just to clarify, if i've a 4 core host cpus can i give 4 cpus for both
> of the 2 guests?
Yes, virtual CPUs are mapped to threads.
Luca
--
s and gcc to emit 16bit code (e.g. HPA
rewrote i386 boot code in plain C)...
Luca
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse012000
IDE probe. Same problem with KVM-41 (last known
working is KVM-36, I haven't tested anything in between). In both
cases -no-kvm-irqchip fixes the problem.
Log attached.
Luca
Linux version 2.6.22.2-42.fc6 ([EMAIL PROTECTED]) (gcc version 4.1.2 20070626
(Red Hat 4.1.2-13)) #1 SMP Wed Aug 15 11
of HAL
> exists what are the advantage/disadvantages them. how can i
> change/configure this?
Hardware Abstraction Layer. See:
http://support.microsoft.com/kb/99588
http://en.wikipedia.org/wiki/Architecture_of_Windows_NT
Luca
---
On 9/21/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Luca wrote:
> > On 9/20/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> >
> >> Not much user visible change besides the OpenBSD regression fix.
> >>
> >> As usual, if you have an issue please tr
nd -serial file:/tmp/log.txt to KVM
command line and add "console=ttyS0" to the guest kernel command line.
Luca
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual S
f [ -x /sbin/ip ]; then
> + switch=$(/sbin/ip route list | awk '/^default / { print $NF }')
NAK on this piece. This won't work with non-trivial setups (yes, I now
that the original script is broken too, but since you are touching
it...).
See:
http://bugs.debian.or
On 12/31/06, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Luca Tettamanti wrote:
> > Hello,
> > I'm testing KVM on a Core2 CPU. I'm running kernel 2.6.20-git (pulled
> > few hours ago), configured with SMP and PREEMPT.
> >
> > I'm hitting 2 di
se cfq as the
> I/O scheduler.
Is cfq still fair in the guest? The VM re-dispatches the requests (at
least when using QEMU IDE) and the host can reschedule them at will.
Luca
-
This SF.net email is sponsored by the 2008 Java
; +static int devfn = 7;
> +
> +if ((devfn % 8) == 7)
> + devfn = -1;
> +else
> + devfn++;
This code look strange... devfn should be passed to virtio_init_pci by
virtio-{net,blk} init functions, no?
Luca
---
lt ACPI script of the debian stops the desktop
manager if it's running but doesn't shut down the machine. If the ACPI
daemon is not running in the guest then the event is lost...
Luca
-
This SF.net email is spo
#x27;s why today some people are hunting copies and submitting patches
to cleanup that mess.
lu
--
Luca Barbato
Gentoo Council Member
Gentoo/linux Gentoo/PPC
http://dev.gentoo.org/~lu_zero
-
This SF.net email is sponsored by:
.1 vanilla has survived and is currently still under test.
> kvm.git tag master killed our filesystem at least three times.since monday.
linus-git has at least one bug with SG chaining, but usually it just
hangs the machine. Patch is here:
http://lkml.org/lkml/2007/10/17/269
Luca
--
m does not help.
Try acpi=force on the kernel command line (guest). QEmu doesn't expose
a DMI block and Linux guests usually don't enable ACPI since they're
unable to check BIOS year (all BIOSes shipped before 2001 are
blacklisted be
was fixed in QEMU CVS a while ago, I believe that Avi pulled that
patch when rebased KVM userspace.
Luca
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Sear
hen using RTC or HPET for timing you also need to mask SIGIO; you may
also need to change F_SETOWN (enable_sigio_timer) with the correct
thread id, i.e.:
fcntl(fd, F_SETOWN, gettid());
Luca
-
This SF.net email is sponsored by: Splunk Inc.
> They'll never be touched by anyone so it doesn't actually increase the
> resident size.
Iff the host kernel is configured to overcommit memory.
Luca
-
SF.Net email is sponsored by:
Check out t
m_user+0x26/0xe6
[] kvm_vcpu_ioctl+0x0/0xcff [kvm]
[] do_ioctl+0x1f/0x62
[] vfs_ioctl+0x237/0x249
[] sys_ioctl+0x33/0x4d
[] syscall_call+0x7/0xb
===
Luca
--
Il coraggio non mi manca.
E` la paura che mi frega...
--
Hi,
Il Mon, Jun 04, 2007 at 12:35:37PM +0300, Avi Kivity ha scritto:
> Luca Tettamanti wrote:
> >Hello,
> >my kernel just exploded :)
> >
> >The host is running 2.6-git-current, with KVM modules from KVM-27
> >package. kernel is 32bit, SMP, with PREEMPT enab
Hello,
I'm testing the new Fedora7 under KVM. As you may know Fedora has
migrated to the new libata drivers.
ata_piix is unhappy with the PIIX IDE controller provided by QEmu/KVM:
libata version 2.20 loaded.
ata_piix :00:01.1: version 2.10ac1
PCI: Setting latency timer of device :00:01.1
nolapic', i.e. disabling the local APIC on
the guest kernel.
I've also tried disabling TSC (notsc) and forcing PIT as the clocksource
(clocksouce=pit clock=pit); neither of them helped.
Luca
--
Il dottore mi ha detto
Il Mon, Jun 11, 2007 at 10:44:45AM +0300, Avi Kivity ha scritto:
> Luca wrote:
> >>
> >>I've managed to reproduce this on kvm-21 (it takes many boots for this
> >>to happen, but it does eventually).
> >
> >Hum, any clue on the cause?
>
> Fro
Il Wed, Jun 13, 2007 at 11:59:25AM +0300, Avi Kivity ha scritto:
> Luca Tettamanti wrote:
> >Il Mon, Jun 11, 2007 at 10:44:45AM +0300, Avi Kivity ha scritto:
> >
> >>Luca wrote:
> >>
> >>>>I've managed to reproduce this on kvm-21 (it t
Il Thu, Jun 14, 2007 at 11:26:29AM +0300, Avi Kivity ha scritto:
> Luca Tettamanti wrote:
> >With GOOD_APIC apic_read_around is a nop, while apic_write_around is a
> >normal write. With !GOOD_APIC apic_write_around writes to the APIC reg
> >using xchg. With !GOO
Il Thu, Jun 14, 2007 at 11:26:29AM +0300, Avi Kivity ha scritto:
> Luca Tettamanti wrote:
> >With GOOD_APIC apic_read_around is a nop, while apic_write_around is a
> >normal write. With !GOOD_APIC apic_write_around writes to the APIC reg
> >using xchg. With !GOO
Il Fri, Jun 15, 2007 at 12:53:24AM +0200, Luca Tettamanti ha scritto:
> Il Thu, Jun 14, 2007 at 11:26:29AM +0300, Avi Kivity ha scritto:
> > Luca Tettamanti wrote:
> > >With GOOD_APIC apic_read_around is a nop, while apic_write_around is a
> > >normal write. With !
pa, new, bytes);
+
+ return X86EMUL_CONTINUE;
}
static unsigned long get_segment_base(struct kvm_vcpu *vcpu, int seg)
Both patches apply to kvm-28 and have been run-time tested with 32bit
guest on 32bit host, with a VMX processor.
If patches look good I'll resubmit with proper cha
Il Sat, Jun 16, 2007 at 10:43:23AM +0300, Avi Kivity ha scritto:
> Luca Tettamanti wrote:
> > Il Fri, Jun 15, 2007 at 12:06:50PM +0300, Avi Kivity ha scritto:
> >
> >>> After a bit of thinking: it's correct but removes an optimization;
> >>> fur
and always perform the writeback stage in the
emulation.
Signed-Off-By: Luca Tettamanti <[EMAIL PROTECTED]>
---
drivers/kvm/x86_emulate.c | 60 +++-
1 file changed, 29 insertions(+), 31 deletions(-)
diff --git a/drivers/kvm/x86_emulate.c b/drive
When writing to normal memory and the memory area is unchanged the write
can be safely skipped, avoiding the costly kvm_mmu_pte_write.
Signed-Off-By: Luca Tettamanti <[EMAIL PROTECTED]>
---
drivers/kvm/kvm_main.c |6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff -
and perform the writeback stage in the emulation unless
it's explicitly disabled (currently push and some 2 bytes instructions
may disable the writeback).
Signed-Off-By: Luca Tettamanti <[EMAIL PROTECTED]>
---
Tested with Fedora7, Solaris10 and WinXP on a 32 bit host with Intel
CPU
1 - 100 of 130 matches
Mail list logo
